基于词频同现与WordNet的图像自动标注改善算法研究

图像自动标注是模式识别与计算机视觉等领域中的重要问题。针对现有图像自动标注模型普遍受到语义鸿沟问题的影响,提出了基于关键词同现的图像自动标注改善方法,该方法利用数据集中标注词间的关联性来改善图像自动标注的结果。此外,针对上述方法不能反映更广义的人的知识以及易受数据库规模影响等问题,提出了基于语义相似的图像自动标注改善方法,通过引入具有大量词汇、包含了人知识的结构化电子词典WordNet来计算词汇间的关系并改善图像自动标注结果。实验结果表明,提出的两个图像自动标注改善方法在各项评价指标上相比以往模型均有所提高。     

图象模糊涟缘检测的改进算法

图象在检测技术是图象处理中最重要的内容之一,且已在图象分析和识别领域中得到广泛的应用。针对图象边缘由模糊性引起的不确定性问题,提出了一种图象模糊边缘检测的改进算法,该算法是道德民确定一个阈值参数,然后根据此阈值参数来定义一个新的隶属函数,从而钭图象转化为等效的图象模糊特征平面,通过在模糊特征平面上进行增强运算,将其转换为空域图象,最后再进行边缘提取,同时还对具有多峰直方图分布图象的模糊边缘检测方法进行推广,仿真结果表明,该算法是有效的。     

A study on vulnerability and presentation attack detection in palmprint verification system

As biometric systems become ubiquitous in the domain of personal authentication, it is of utmost importance that these systems are secured against attacks. Among various types of attacks on biometric systems, the presentation attack, which involves presenting a fake copy (artefact) of the real biometric to the biometric sensor to gain illegitimate access, is the most common one. Despite the serious threat posed by these attacks, not much work has been done to address this vulnerability in palmprint-based biometric systems. This paper demonstrates the vulnerability of a palmprint verification system to presentation attacks and proposes a novel presentation attack detection (PAD) approach to discriminating between real biometric samples and artefacts. The proposed PAD approach is inspired by a work that established relationship between the surface reflectance and a set of statistical features extracted from the image. Specifically, statistical features computed from the distributions of pixel intensities, sub-band wavelet coefficients and the grey-level co-occurrence matrix form the original feature set, and CFS-based feature selection approach selects the most discriminating feature subset. A trained binary classifier utilizes the selected feature subset to determine whether the acquired image is of real hand or an artefact. For performance evaluation, an antispoofing database—PALMspoof has been developed. This database comprises left- and right-hand images of 104 subjects, and three kinds of artefacts generated from these images. In addition to PALMspoof database, the biometric system’s vulnerability has been assessed on display and print artefacts generated from two publicly available palmprint datasets. Our experimental results show that 1) the palmprint verification system is highly vulnerable with spoof acceptance of 84.56%; 2) the proposed PAD approach is effective against both print and display attacks, in both same-device and cross-device scenarios; and 3) the proposed approach for PAD provides an average improvement of 12.73 percentage points in classification error rate over local binary pattern (LBP)-based PAD approach.     

Correlation‐based classifier combination in the field of pattern recognition

Classifier combination methods have proved to be an effective tool to increase the performance of classification techniques that can be used in any pattern recognition applications. Despite a significant number of publications describing successful classifier combination implementations, the theoretical basis is still not matured enough and achieved improvements are inconsistent. In this paper, we propose a novel statistical validation technique known as correlation‐based classifier combination technique for combining classifier in any pattern recognition problem. This validation has significant influence on the performance of combinations, and their utilization is necessary for complete theoretical understanding of combination algorithms. The analysis presented is statistical in nature but promises to lead to a class of algorithms for rank‐based decision combination. The potentials of the theoretical and practical issues in implementation are illustrated by applying it on 2 standard datasets in pattern recognition domain, namely, handwritten digit recognition and letter image recognition datasets taken from UCI Machine Learning Database Repository ( http://www.ics.uci.edu/_mlearn ). 1 An empirical evaluation using 8 well‐known distinct classifiers confirms the validity of our approach compared to some other combinations of multiple classifiers algorithms. Finally, we also suggest a methodology for determining the best mix of individual classifiers.     

基于AutoML的保护区物种识别

随着我国在生态保护上的投入加大,红外相机技术在我国各级自然保护区的应用发展迅猛,在如何充分挖掘照片的信息方面,物种识别显得尤为重要,是其他工作的前提.在图像识别方面,随着深度学习的爆发,给图像识别带来了革命性的提升,以卷积神经网络为代表的网络结构在准确率上几乎完胜传统方法.然而,由于网络结构对最终图像识别准确率的影响巨大,人们在实际应用中往往都是使用一些经典的网络结构,比如VGG16、VGG19、ResNet50等,从中选择一个适合自己的数据集的网络结构,同时对于不同的数据集,可能需要重新选择.因此,在保护区红外相机物种的识别中,本文提出了基于AutoML的自动构建网络结构技术,针对不同的保护区的数据集,自动构建合适的网络结构,避免人工选择,同时达到了与人工选择网络相当的准确率.     

基于判别模型与生成模型的层叠图像自动标注

图像自动标注是模式识别与计算机视觉等领域中重要而又具有挑战性的问题.针对现有模型存在数据利用率低与易受正负样本不平衡影响等问题,提出了基于判别模型与生成模型的新型层叠图像自动标注模型.该模型第一层利用判别模型对未标注图像进行主题标注,获得相应的相关图像集;第二层利用提出的面向关键词的方法建立图像与关键词之间的联系,并使用提出的迭代算法分别对语义关键词与相关图像进行扩展;最后利用生成模型与扩展的相关图像集对未标注图像进行详细标注.该模型综合了判别模型与生成模型的优点,通过利用较少的相关训练图像来获得更好的标注结果.在Corel 5K图像库上进行的实验验证了该模型的有效性.     

食品图像识别方法综述

食品与人类的行为、健康和文化等密切相关.社交网络、移动网络和物联网等泛在网络产生了食品大数据,这些大数据与人工智能,尤其是快速发展的深度学习催生了新的交叉研究领域食品计算.作为食品计算的核心任务之一,食品图像识别同时是计算机视觉领域中细粒度视觉识别的重要分支,因而具有重要的理论研究意义,并在智慧健康、食品智能装备、智慧...     

Pose-specific non-linear mappings in feature space towards multiview facial expression recognition

We introduce a novel approach to recognizing facial expressions over a large range of head poses. Like previous approaches, we map the features extracted from the input image to the corresponding features of the face with the same facial expression but seen in a frontal view. This allows us to collect all training data into a common referential and therefore benefit from more data to learn to recognize the expressions. However, by contrast with such previous work, our mapping depends on the pose of the input image: We first estimate the pose of the head in the input image, and then apply the mapping specifically learned for this pose. The features after mapping are therefore much more reliable for recognition purposes. In addition, we introduce a non-linear form for the mapping of the features, and we show that it is robust to occasional mistakes made by the pose estimation stage. We evaluate our approach with extensive experiments on two protocols of the BU3DFE and Multi-PIE datasets, and show that it outperforms the state-of-the-art on both datasets.     

Directed enumeration method in image recognition

The article is devoted to the problem of image recognition in real-time applications with a large database containing hundreds of classes. The directed enumeration method as an alternative to exhaustive search is examined. This method has two advantages. First, it could be applied with measures of similarity which do not satisfy metric properties (chi-square distance, Kullback–Leibler information discrimination, etc.). Second, the directed enumeration method increases recognition speed even in the most difficult cases which seem to be very important in practical terms. In these cases many neighbors are located at very similar distances. In this paper we present the results of an experimental study of the directed enumeration method with comparison of color- and gradient-orientation histograms in solving the problem of face recognition with well-known datasets (Essex, FERET). It is shown that the proposed method is characterized by increased computing efficiency of automatic image recognition (3–12 times in comparison with a conventional nearest neighbor classifier).     

A rule-based system for automatic seismic discrimination

This paper proposes a rule-based system for automatic seismic discrimination, i.e. classification of earthquakes and underground nuclear explosions. It incorporates rule-based deduction, pattern recognition and signal processing for an effective identification. The seismological knowledge and heuristics are represented by a set of production rules. Facts and assertions of the production rules are derived from seismic signals using signal processing and pattern recognition methods. Due to the uncertainty nature of this problem there are certainly factors associated with both antecedents and the rules. The control strategy is data-driven, i.e. forward-chaining for better efficiency. This approach can be applied to other signal and image interpretation problems.     

Unsupervised building detection in complex urban environments from multispectral satellite imagery

A generic algorithm is presented for automatic extraction of buildings and roads from complex urban environments in high-resolution satellite images where the extraction of both object types at the same time enhances the performance. The proposed approach exploits spectral properties in conjunction with spatial properties, both of which actually provide complementary information to each other. First, a high-resolution pansharpened colour image is obtained by merging the high-resolution panchromatic (PAN) and the low-resolution multispectral images yielding a colour image at the resolution of the PAN band. Natural and man-made regions are classified and segmented by the Normalized Difference Vegetation Index (NDVI). Shadow regions are detected by the chromaticity to intensity ratio in the YIQ colour space. After the classification of the vegetation and the shadow areas, the rest of the image consists of man-made areas only. The man-made areas are partitioned by mean shift segmentation where some resulting segments are irrelevant to buildings in terms of shape. These artefacts are eliminated in two steps: First, each segment is thinned using morphological operations and its length is compared to a threshold which is determined according to the empirical length of the buildings. As a result, long segments which most probably represent roads are masked out. Second, the erroneous thin artefacts which are classified by principal component analysis (PCA) are removed. In parallel to PCA, small artefacts are wiped out based on morphological processes as well. The resultant man-made mask image is overlaid on the ground-truth image, where the buildings are previously labelled, for the accuracy assessment of the methodology. The method is applied to Quickbird images (2.4 m multispectral R, G, B, near-infrared (NIR) bands and 0.6 m PAN band) of eight different urban regions, each of which includes different properties of surface objects. The images are extending from simple to complex urban area. The simple image type includes a regular urban area with low density and regular building pattern. The complex image type involves almost all kinds of challenges such as small and large buildings, regions with bare soil, vegetation areas, shadows and so on. Although the performance of the algorithm slightly changes for various urban complexity levels, it performs well for all types of urban areas.     

基于类别方差的自动多门限图象分割

从模式识别的分类理论出发,对基于类别方差的自动单门限图象分割方法进行了扩展,给出一种适合于灰度图象的自动多门限图象分割方法,对灰度图象能够自动寻找最优的门限值（一个或多个）,与人类的视觉分割过程一致。实验表明,能够自动且有效地进行灰度图象的多级门限值分割。     

A two-level model for automatic image annotation

Image automatic annotation is a significant and challenging problem in pattern recognition and computer vision. Current image annotation models almost used all the training images to estimate joint generation probabilities between images and keywords, which would inevitably bring a lot of irrelevant images. To solve the above problem, we propose a hierarchical image annotation model which combines advantages of discriminative model and generative model. In first annotation layer, discriminative model is used to assign topic annotations to unlabeled images, and then relevant image set corresponding to each unlabeled image is obtained. In second annotation layer, we propose a keywords-oriented method to establish links between images and keywords, and then our iterative algorithm is used to expand relevant image sets. Candidate labels will be given higher weights by using our method based on visual keywords. Finally, generative model is used to assign detailed annotations to unlabeled images on expanded relevant image sets. Experiments conducted on Corel 5K datasets verify the effectiveness of our hierarchical image annotation model.     

OPTIMOL: Automatic Online Picture Collection via Incremental Model Learning

The explosion of the Internet provides us with a tremendous resource of images shared online. It also confronts vision researchers the problem of finding effective methods to navigate the vast amount of visual information. Semantic image understanding plays a vital role towards solving this problem. One important task in image understanding is object recognition, in particular, generic object categorization. Critical to this problem are the issues of learning and dataset. Abundant data helps to train a robust recognition system, while a good object classifier can help to collect a large amount of images. This paper presents a novel object recognition algorithm that performs automatic dataset collecting and incremental model learning simultaneously. The goal of this work is to use the tremendous resources of the web to learn robust object category models for detecting and searching for objects in real-world cluttered scenes. Humans contiguously update the knowledge of objects when new examples are observed. Our framework emulates this human learning process by iteratively accumulating model knowledge and image examples. We adapt a non-parametric latent topic model and propose an incremental learning framework. Our algorithm is capable of automatically collecting much larger object category datasets for 22 randomly selected classes from the Caltech 101 dataset. Furthermore, our system offers not only more images in each object category but also a robust object category model and meaningful image annotation. Our experiments show that OPTIMOL is capable of collecting image datasets that are superior to the well known manually collected object datasets Caltech 101 and LabelMe.     

Foam-rubber sample volume measuring

For industrial quality control of foam-rubber material, it is required to measure volume of the sample. A new approach is proposed to measure sample volume by images of sample faces. Faces images are got via flatbed scanner. The faces images are processed and the sample is approximated by hexahedron. Then the sample volume is calculated analytically. Also we proposed an iterative approach based on splitting geometrical model of the sample into several smaller hexahedrons. The test results have shown that results of volume measurements obtained by proposed approach coincide well with ones obtained by the standard method. However, repeatability and reproducibility of measurements is better for proposed algorithm, and it is faster. The article is published in the original. Ilia V. Safonov. Received his MS degree in automatic and electronic engineering from Moscow Engineering Physics Institute/University (MEPhI), Russia in 1994 and his PhD degree in computer science from MEPhI in 1997. Since 1998 he is an associate professor of faculty of Cybernetics of MEPhI while conducting researches in image segmentation, features extraction and pattern recognition problems. Since 2004, Dr. Safonov has joint Image Enhancement Technology Group, Printing Technology Lab, Samsung Research Center, Moscow, Russia where he is engaged in photo, video and document image enhancement projects. Sergei Yu. Yakovlev. Received the MS degree in cybernetics from Moscow Engineering Physics Institute/University (MEPhI), Russia in 2005. He is presently working towards his PhD degree. His current research interests include image processing, pattern recognition and 3D shape reconstruction.     

基于训练样本自动选取的SVM彩色图像分割方法

图像分割是模式识别、图像理解、计算机视觉等领域的重要研究内容。基于支持向量机((Support Vcctor Ma- chine, SVM)的方法现已广泛应用于图像分割,但其在训练样本的选取上大多是人工选择,这降低了图像分割的自适 应性,且影响了SVM的分类性能。提出一种基于训练样本自动选取的SVM彩色图像分割方法,算法首先使用模糊 C均值(Fuzzy C-Mcans, FCM)聚类算法自动获取训练样本,然后分别提取图像颜色特征和纹理特征,将其作为SVM 模型训练样本的特征属性进行训练,最后用训练好的分类器对图像进行分割。实验结果表明,提出的方法可取得很好 的分割结果。     

视频火灾检测的设计与实现

在分析火灾图像特性的基础上,运用数字图像处理技术和模式识别技术,提出了火灾识别的思想.给出了图像处理和识别的算法,该算法采用二维最大熵自动阈值法对火灾图像进行分割处理,分割后再提取可疑区域;对可疑区域的火焰进行识别,给出火焰存在的可能性;根据火灾火焰蔓延时的面积、相似度的变化来识别、判断火灾的发生.实验证明,与传统的检测方法相比,大大地提高火灾预报的准确率.     

场景文字识别算法的研究

文字识别是一种通用的图像理解技术,对信息检索、车牌识别和自动驾驶等应用的研究有着重要意义。随着神经网络的伟大复兴,场景文字识别任务得到了很大推动,近年来涌现了许多基于深度学习的文字识别算法。本文提出了一种基于特征融合的CRNN改进算法,使用三个通用的文字识别数据集从识别准确率、运行效率和模型大小三个方面进行分析。实验结果表明该算法在提高准确率的同时,运行效率也有所提高。     

多姿态人脸识别

人脸识别在很多场合都有重要的作用,传统的身份验证是采用某种识别号码等方法, 以阻止伤造的发生。由于人的视觉特征如面部,姿态等是相对稳定而且各不相同的,因此采用这些特征进行身份的识别是可行的本文提出了一种处理多姿态人脸识别的多候选类加权识别方法,为了减少姿态变化的影响,提出了相应的预处理法。